# Detection of Pesticide Residues Using Three-Dimensional SERS Substrate Based on CNTs/Ag/AgNWs/SiO2

## TL;DR

This study introduces a 3D SERS substrate that improves detection of pesticide residues with high sensitivity and accuracy.

## Contribution

A novel 3D CNTs/Ag/AgNWs/SiO2 SERS substrate with enhanced sensitivity and a GA-BP algorithm for pesticide detection.

## Key findings

- The substrate achieved a detection limit of 10−12 mol/L for R6G and an enhancement factor of 8.947 × 10^8.
- The detection limit for thiram was 0.1 mg/L, below regulatory standards in China and the EU.
- The GA-BP algorithm achieved 92.5% prediction accuracy for thiram concentrations.

## Abstract

In response to the shortcomings of traditional surface-enhanced Raman spectroscopy (SERS) substrates, such as short shelf life, poor uniformity, and low selectivity, this study innovatively proposed a three-dimensional composite substrate of CNTs/Ag/AgNWs/SiO2. This substrate demonstrates excellent SERS enhancement effects, with a detection limit of 10−12 mol/L for the probe molecule Rhodamine 6G (R6G) and an enhancement factor (EF) of 8.947 × 108. Further experiments confirmed the substrate’s superior uniformity and stability. The enhancement mechanism was investigated using both experimental methods and the Finite Difference Time Domain (FDTD) approach. When commonly used pesticide thiram was used as the target analyte, the detection limit of the substrate reached 0.1 mg/L, which is significantly lower than the pesticide residue standards of China and the European Union. Additionally, the genetic algorithm (GA)-optimized Back Propagation (BP) neural network was introduced for the quantitative analysis of thiram concentrations. The experimental results indicated that the GA-BP algorithm achieved the training prediction accuracy of 92.5% for thiram, demonstrating good network performance. This method shows good selectivity and has broad application prospects in the detection of toxic chemicals, environmental pollutants, and food additives.